Description du projet
Améliorer la fiabilité et la latence des réseaux d’intelligence artificielle et d’apprentissage automatique
La communication à faible latence joue un rôle de plus en plus crucial avec l’apparition d’applications sensibles à la latence comme les réseaux et usines intelligents et l’Internet des objets. Les communications à haute fiabilité et faible latence (HRLL pour «High-reliability low-latency») visent à répondre aux immenses exigences de nombreuses applications en termes de fiabilité et de latence. Le codage réseau est susceptible d’améliorer la fiabilité, la latence et les débits de données. Cela dit, les recherches sur le codage réseau destiné aux communications HRLL font cruellement défaut. Le projet RACOON, financé par l’UE, entend mener des études pour déterminer les limites clés des réseaux HRLL à l’aide du codage réseau. Il optimisera les systèmes de codage réseau et proposera des approches de communication efficaces qui amélioreront la latence et la fiabilité futures des applications critiques.
Objectif
With various emerging latency-sensitive applications, e.g. V2V (vehicle-to-vehicle) automation, smart power systems (i.e. smart grids), Internet of things (IoT) and smart factory (industry 4.0) low-latency communication is becoming more and more important. To meet increasing demands, there have been substantial research activities on low-latency communications recently. Meanwhile, many applications also have high requirements on reliability. These schemes with high requirements on reliability and latency are often termed high-reliability low-latency (HRLL) communications. A common approach to achieve low-latency is finite length coding/transmission blocks. However, due to finite code length, it is rather challenging to achieve high reliability and low latency simultaneously. Moreover, in many scenarios, high data rates are also needed. Thus, it is rather valuable to study communication and coding schemes capable of simultaneously improving reliability, latency and rates. Network coding, originally proposed for increasing network throughput, has shown benefits in latency, throughput and reliability, respectively. However, the study of network coding for HRLL communications is still very limited. Thus, the main objective of the project is to systematically investigate network coding for HRLL communications. We will find the fundamental limits of HRLL networks with network coding. We will optimize network coding schemes and propose efficient communication strategies to improve latency and reliability performance under practical constraints. The outgoing mobility will be hosted Electrical Engineering department, Princeton University, USA, with excellent academy environments and the applicant will be trained in advanced knowledge, management and pedagogy skills, in addition to research topics and methods.
Champ scientifique
Programme(s)
Régime de financement
MSCA-IF-GF - Global FellowshipsCoordinateur
100 44 Stockholm
Suède